First discovery of an unpopulated land can be a matter of personal and national pride and may also form the basis of territorial claims. In the case of South Georgia, first discovery is disputed, as is sovereignty. The island was claimed for Britain in 1775 (as the Isle of Georgia) and for Argentina (as Isla San Pedro) in 1927 (Palazzi, Reference Palazzi2005, p. 232). The United Kingdom offered to take the dispute about its Antarctic and sub-Antarctic territories (the Falkland Islands Dependencies) to the International Court of Justice in 1947 and 1948, but Argentina declined. In 1955, as resolution of such disputes became important in the negotiations that were to lead to the Antarctic Treaty, the United Kingdom instituted unilateral proceedings in the ICJ with a view to having its claim endorsed, noting in its application that

In summing up this section, the Government commented that “the first discoveries of South Georgia, the South Sandwich Islands, the South Orkneys, the South Shetlands. and Graham Land were thus all made by British nationals …. There were no Spanish or Argentine discoveries” (International Court of Justice, 1956, p. 12). In the event, Argentina refused to accept the Court’s jurisdiction to deal with the case, so the application did not proceed. Two points in this part of the text are striking. First, there is an inconsistency between “may well have been” (“il est très possible”, in the French text) and “rediscovered in 1775”. Second, there is no mention of the intervening claim to first discovery in 1756 by the Spanish navigator Gregorio Jerez (of whom more below).

The purpose of this note is not to assess the rival claims to sovereignty, but rather to investigate the claim that South Georgia may have been first discovered in April 1675 by an English merchant called Antonio (or Antoine or Antonie or Anthony) de la Roche (or La Roché), whose surname was that of his French father. La Roche’s primacy is widely regarded a probability rather than a fact. Robert Headland, for example, says in his The Island of South Georgia that the first discovery was “most probably the sighting of South Georgia in 1675 by a London merchant, Antoine de la Roché” (Headland, Reference Headland1984, p. 21). In Headland’s authoritative Chronology of Antarctic Exploration the claim is softened slightly: La Roche is said to have “probably discovered South Georgia April 1675, while blown off course”, and “details of the voyage remain unclear” (Headland, Reference Headland2009. s.v. 1674–1675). There are two unresolved questions: whether La Roche was the first to see the island, and whether the island that he saw was South Georgia. I should like in this note to introduce a third uncertainly, by questioning whether Antonio de la Roche ever existed. I shall review the traditional answers to the first two questions and approach the third through my own archival investigations and by drawing on recent research on the sources of the earliest accounts of La Roche.

The contention that La Roche was the probably the first discoverer of South Georgia was long contested by those supportive of the rival claim for Amerigo Vespucci, whose voyages are not well attested. The voyage on which some believe he sighted South Georgia occurred in 1500–1501, when Vespucci travelled in an unknown capacity on a voyage to the western Atlantic commissioned by Manuel I of Portugal, probably with a view to ascertaining where the landmass recently discovered by Pedro Álvares Cabral lay in relation to the line of demarcation established in the Treaty of Tordesillas.

The historicity of this voyage is confirmed by a variety of sources, but the sole source of details of the route is contained in a letter of 1504 said to have been written by Vespucci to Piero Soderini, head (gonfaloniere) of the government of Florence. The authenticity of the letter has long been debated, but in the authoritative judgement of Felipe Fernández-Armesto, the letter is a confection by other hands with some passages that seem to have originated with Vespucci (Fernández-Armesto, Reference Fernández-Armesto2006, pp. 126–135). The letter is complex, in part because of the mixture of Italian, Spanish and Portuguese forms, but also because it exists in three manuscript versions (with textual variants) in Italian and one printed version in Latin (see Formisano, Reference Formisano2020). The latter, known as the Hylacomyus version, is printed as Quatuor Americi Vespuccij navigationes in Martin Waldseemüller’s Cosmographiae Introductio (1507).

Vespucci’s voyage as described in the Soderini letter began with a crossing to the coast of South America at 5˚ S. The expedition then turned southwards to the Baía do Trapandé at 25˚ S (close to the longitude through which the Tordesillas Line passed) and then sailed southwest as far as 52˚ S, whereupon on 7 April new land was sighted: avemmo vista di nuovo terra della quale coremmo circa di venti leghe e la trovammo tutta costa brava o bizarre. The meaning of the final phrase is perhaps clearer in the Latin text (brutalem et extraneam), so the sentence might be translated “we sighted new land, travelling along it for some 20 leagues, and found it to be a wild and alien coast”. The description and location are so vague that it could apply to any number of islands, including the Falklands or Isla de los Estados (English exonym: Staten Island) or Tristan de Cunha as well as South Georgia. The evidence was carefully reviewed by in 1950 by E.W.H. Christie, who concluded that there is “no reason whatsoever for naming Amerigo Vespucci as the discoverer of South Georgia” (Christie, Reference Christie1950). As for La Roche, Christie argued that “there is every reason for the impartial observer to accept the view that the snow-covered mountainous land seen by La Roché was the island of South Georgia” (Christie, Reference Christie1951, p. 41).

The sole source of information on La Roche is the writing of the Spanish mariner and writer Francisco de Seyxas y Lovera (c. 1646–c. 1705), who had described La Roche’s journey in Descripcion geographica, y derrotero de la region austral magallanica, published in Madrid in 1690. Seyxas explains that in 1674 La Roche acquired a 350-ton ship in Hamburg, secured a license to trade in Spanish America and proceeded south accompanied by a 50-ton bilander. In May 1674, he called at the Canary Islands to take on supplies, and then sailed south, travelling through the Straits of Magellan and arriving in Callao (Peru) in October. On the return journey, he stopped in Isla de Chiloé (Chile) to careen the two ships, and then set out for Bahia de Todos os Santos (Brazil). He failed to enter the Straits of Magellan and so rounded the Cape. He encountered bad weather and was unable to enter the Le Maire Strait that separates the eastern tip of Tierra del Fuego from Isla de los Estados. He attempted to round Isla de los Estados to the south in order to enter the (mythical) “Brouwer’s Strait” shown on Dutch maps, but was blown far to the east, where he sighted land. Sailing closer, they found

When the storm abated, La Roche sailed around the southeast tip of the island and then resumed his journey, discovering another unknown island (Isla Grande) before landing at Bahia de Todos os Santos, and eventually returning to Europe, arriving at the French port of La Rochelle on 29 September 1675.

The island that La Roche had discovered soon began to appear on maps. The first was a map of the Région Austral Magallánica that in 1692 Seyxas added to “Atlas Taboas geraes da toda a navegacão” (1630), a manuscript atlas (now in the Library of Congress, G1015.T4 1630) by the Portuguese cartographer João Teixeira Albernaz I (McCarl, Reference McCarl2018, Reference McCarl2020). In 1703, the French cartographer Guillaume Delisle produced a Carte du Paraguay, du Chili, du Detroit de Magellan, which showed Ile de La Roche divided by a strait from a piece of land to the east labelled “terre inconnüe”; the strait is labelled “Detroit de la Roche découvert en 1675” (Delisle, Reference Delisle1703). This image of Ile de la Roche was soon to be borrowed by mapmakers in Amsterdam (Henri Chatelain; Chatelaine, Reference Chatelain1705), Paris (Nicolas de Fer; De Fer, Reference De Fer1720; Philippe Buache; Buache, Reference Buache1754), Nuremburg (Homann Heirs, 1733), Venice (Giovanni Battista Albrizzi; Albrizzi, Reference Albrizzi1740) and London (Herman Moll;. Moll, c, Reference Moll1709. Charles Price; Price, c. Reference Price1710; John Lodge Cowley; Cowley, Reference Cowley and Hübner1753; Richard William Seale; Seale, c. Reference Seale1745). On all these maps, Roche Island is placed closer to the South American mainland than the longitude of the island that we now know as South Georgia, with which Roche Island was later to be identified.

In 1756, cartography began to be overtaken by events. A Spanish merchant vessel called León (registered as Santo Christo del Auxilio y Nuestra Señora de los Dolores), commanded by Gregorio Jerez, sailed south from Callao (Peru) and rounded the Cape. One of the passengers on board was Nicolas-Pierre Duclos-Guyot, who recorded being blown east in a severe storm. On 30 June, land was sighted, and a reading at noon established its position to be 54˚50’S and 51˚32’W. The island was named Ⓘle de St Pierre in French and Isla de San Pedro in Spanish. La Roche was not mentioned. The coordinates taken by Gregorio Jerez were not particularly accurate, but nonetheless make clear that what had been found was the island now known in the Anglophone world as South Georgia. The account of Duclos-Guyot meant that Roche Island was placed much further east than on previous maps. “I de S Pierre découv. en 1756” soon appeared on the maps of Jean-Baptiste Bourguignon d’Anville (Geographer to Louis XVI). His maps continued to credit the Spanish expedition with first discovery for decades. A “Map of South America from M^r D’Anville” published in English by Laurie and Whittle in 1794 declares that the “I[sle] of Georgia” was “discovered in 1756 by the Spanish ship Lion and explored in 1775 by Cap^t Cook” (Bourguignon d’Anville, Reference Bourguignon d’Anville1794). Again, La Roche is not mentioned. In 1769, the Scottish geographer and hydrographer Alexander Dalrymple published a Chart of the Southern Ocean (Dalrymple, Reference Dalrymple1769a) together with a pamphlet called Memoir of a chart of the Southern Ocean (Dalrymple, Reference Dalrymple1769b). The chart, engraved by Thomas Jeffreys [sic], shows the Str[ait] de la Roche separating an unnamed island from a large land mass, the edge of the continent that Dalrymple believed existed.

When Commander Cook (not yet promoted to Captain), on his second voyage, rounded Cape Horn on 29 December 1774 (Cook, Reference Cook and Edwards1999, p. 405), he was carrying on board HMS Resolution a copy of Dalrymple’s chart and another by Jean-Baptiste d’Anville. Cook assumed that Ⓘle de St Pierre and Roche Island were different names for the same island, which he wished to find. He adjusted his course,

On 16 January, Cook sighted the land for which he had been searching and proceeded to survey it (Cook, Reference Cook and Edwards1999, pp. 407-08). Having established that the land was an island, Cook named it the Isle of Georgia (after King George III) and on 27 January recorded in his journal his doubt that

Cook’s characteristically nuanced judgement was not shared by all those on board, including James Burney, who was serving as an able seaman.

Burney had read Dalrymple’s Memoir, which had introduced to English readers the name of Francisco de Seyxas y Lovera. Dalrymple subsequently published a translation of Seyxas’ account in A collection of voyages chiefly in the southern Atlantick ocean (Dalrymple, Reference Dalrymple1775). He later read the account of La Roche by Charles Pierre Claret, Comte de Fleurieu, King Louis XVI’s Minister of the Navy and the Colonies (Ministre de la Marine et des Colonies), who in 1797 had published a set of Notes géographiques et historiques appended to an account of the voyage of Jean François de Galaup, comte de Lapérouse (Milet-Mureau, Reference Milet-Mureau1797). Dalrymple and Fleurieu had translated Seyxas’ account of La Roche into English and French, respectively.

In 1813, Burney, by now promoted to Captain, published the third volume of his magnum opus, A Chronological History of the Discoveries in the South Sea or Pacific Ocean (Burney 1803–1817). In chapter 15, the ‘Voyage of Antonio de la Rochè’, Burney assembles all the evidence necessary to evaluate Seyxas’ account of La Roche. He notes that Seyxas was the sole authority for several voyages, including those of John Templemant and Thomas Peche, as well as that of La Roche. He prints an English translation of Seyxas’s account of La Roche in Descripcion geographica, y derrotero de la region austral magallanica (Seyxas, Reference Seyxas y Lovera1690), in which Seyxas explains that his source was a pamphlet ‘privately printed in London, in 12 sheets bound in quarto, in the year 1678, and in the French language’. Burney then sets out ‘the difficulties, conjectures and doubts that have arisen respecting the discoveries of La Rochè’, judiciously comparing the relative merits of the translations of Dalrymple and Fleurieu. He is troubled that Cook could not find the land to the south or south-east of South Georgia that Seyxas had described, and perplexed that the ‘Isla Grande’ that La Roche was supposed to have discovered when returning to South America could not now be found, and tentatively concluded that ‘there does not exist such an island’. He was content, however, to identify Roche Island with South Georgia.

The notion that Amerigo Vespucci had preceded La Roche still lived on. The French naval officer and hydrologist Louis-Isidore Duperry concluded that

This opinion was endorsed by Alexander von Humboldt, who deferred to the argument of “mon respectable ami, M. Duperry” (Humboldt, Reference Humboldt1836–1839, tome 5, p 109)

In 1931, L. Harrison Matthews (Matthews, Reference Matthews1931) used Seyxas’s account of La Roche’s soundings in the inlet in which he was sheltering and of the glaciated mountains that surround it to be identified with Drygalski Fjord. Certainly, Drygalski Fjord affords the appropriate shelter, as I can attest from my own experience in the Fjord as a passenger on a commercial expedition ship. A high land covered with snow sighted to the southeast was tentatively deemed by Matthews to be Clerke Rocks, and the ’snow’ identified as guano, at least in part. The historicity of La Roche’s discovery was cautiously endorsed in the naming of Roché Peak (on Bird Island, off the coast of South Georgia) by the UK Antarctic Place Names Committee in 1960. The British Antarctic Survey Report on The history of place-names in the Falkland Island Dependencies comments that the peak “was named after Antonio de la Roché (or Antoine de la Roche), a London merchant of French parentage, who probably discovered South Georgia in 1675” (Hattersley-Smith, Reference Hattersley-Smith1980, p. 73). The Bulgarian Antarctic Place-Names Commission was less cautious. Naming Roché Glacier (ледник Роше) in Vinson Massif (on the mainland of Antarctica) and explaining in its narrative citation that that the glacier is “named after the British pioneer of Antarctica Anthony de la Roché who discovered the first land in the Antarctic region (Roché Island, present South Georgia) in 1675”; the name was recorded by the International Science Council’s Scientific Committee on Antarctic Research in 2010. In 2019, the video game “Elite Dangerous” featured “Anthony de la Roche Terminal”, a space station in the fictional star system of Badar.

The authenticity of La Roche’s claim has long been questioned, beginning with Burney. There is, for example, a scholarly entry on ‘Antonio de la Roché’ in the Dictionary of Falklands Biography (David, Reference David and Tatham2008), written by Lieutenant-Commander Andrew David, R.N. (former Vice-President of the Hakluyt Society). Commander David’s doubts originate in the account of Isla Grande, on which La Roche said that he had stayed for six days while returning to England. Dalrymple’s chart shows a large island due north of ‘Roché Island’ at 45˚ South. This island was marked on charts for many years thereafter, but on a pocket globe of 1834 now in the National Maritime Museum in Greenwich (GLB0011.1) it is marked ‘very doubtful’ and was eventually agreed not to exist. This troubling fact had led the naval historian Lieutenant-Commander Rupert Gould to think that La Roche need not have been lying, because at 45˚ South on the South American mainland, ‘there are two projecting headlands, either of which can easily be mistaken for an island’ (Gould, Reference Gould1928, p. 203). Commander David concludes that if La Roche’s understanding of his latitude was so far amiss in the case of Isla Grande, then one must question the authenticity of his discovery of South Georgia. On this charitable reading, La Roche may have been merely mistaken. A similar view is taken by the Argentine scholar Ricardo Capdevila (Reference Capdevila1988), who argues that La Roche had not anchored in South Georgia, but in the Falklands.

It is possible to extend that doubt to the question of whether La Roche actually existed. There were two groups of French churches in late seventeenth-century London. The mother churches were at Threadneedle Street, which was Calvinist, and the French Church of the Savoy, which worshipped with a French translation of the Book of Common Prayer. In the late seventeenth century, these churches repeatedly split and regrouped as new churches; by the end of the century, there were some 20 French Protestant churches in London, nine of them in Spitalfields. Huguenot refugees in London and their descendants can normally be found in the records of these two clusters of churches.

I have searched the registers of London’s Huguenot churches: the Chapel Royal in St James Palace (Minet & Minet, Reference Minet and Minet1924), the Church of the Savoy (Minet & Minet, Reference Minet and Minet1922), the Church of the Tabernacle in Milck Alley (Minet & Minet, Reference Minet and Minet1926), the Church of the Artillery in Spitalfields (Minet, Reference Minet1948), the Church of St Jean in Spitalfields (Minet, Reference Minet1938), the Church of St Martin Orgars in Martin Lane (Minet & Minet, Reference Minet and Minet1935), the Church of La Patente de Soho (Minet, Reference Minet1956), the church known as La Patente in Spittlefields (Waller, Reference Waller1898), the Church of Le Petit Charenton in Newport Market (Minet & Minet, Reference Minet and Minet1929) and the churches known by their street names or areas: Berwick Street (Minet & Minet, Reference Minet and Minet1921), Crispin Street (Minet & Minet, Reference Minet and Minet1929), Glasshouse Street (Minet & Minet, Reference Minet and Minet1926), Hoxton (Minet, 1926), Hungerford Market, later Castle Street (Minet & Minet, Reference Minet and Minet1928), Leicester Fields (Minet & Minet, Reference Minet and Minet1926), Pearl Street (Minet & Minet, Reference Minet and Minet1929), Rider Court (Minet & Minet, Reference Minet and Minet1927), Spring Gardens (Minet & Minet, Reference Minet and Minet1922), Swallow Street (Minet & Minet, Reference Minet and Minet1924), Swanfields (Minet & Minet, Reference Minet and Minet1924), Threadneedle Street (Moens, Reference Moens1896, Reference Moens1899; Colyer-Fergusson, Reference Colyer-Fergusson1906, Reference Colyer-Fergusson1916), West Street (Minet & Minet, Reference Minet and Minet1929) and Wheeler Street (Minet, Reference Minet1956). In the case of the Threadneedle Church, I have also searched the Book of Testimonies (Livre des Tesmoignages), (Minet & Minet, Reference Minet and Minet1909), the minutes of the Consistory (Actes du Consistoire) (Gwynne, Reference Gwynn1994; Johnston & Oakley, Reference Johnston and Oakley1937; Oakley, Reference Oakley1969) and the Calendars of the Letter Books (Gwynne, Reference Gwynn1979). In the case of the Church of the Savoy, I have also searched the Book of Conversions and Recognitions (Livre des Conversions et des Reconnoissances) (Minet & Minet, Reference Minet and Minet1914). Finally, I have searched the acts of the colloquies of the French churches (Chamier, Reference Chamier1890), the acts of the synods of the foreign churches (Chamier, Reference Chamier1890) and the indexes of Huguenot wills and administrations (Wagner & North, Reference Wagner and North2007). There are several individuals surnamed Roche and La Roche, but no man of that name has the given name Antoine or any of its variants. I have read the family files on ‘Roche’ and ‘La Roche’ in the Huguenot family research files housed in the National Archives at Kew and again have not found the name Antoine. Of course absence of evidence does not constitute evidence of absence, but one would normally expect to find an historical figure named somewhere in these records.

La Roche does not appear in the historical record, nor does his writing. Despite extensive searches (Christie, Reference Christie1951, p. 41), no copy of the pamphlet La Roche is said to have published in London in 1678 has ever been found. No copy appears in the catalogues of research libraries in the UK or France or the USA, in the catalogues of private British collections such as the Royal Library and the libraries of National Trust properties, or in global catalogues such as WorldCat. That means that Seyxas is the sole authority for La Roche’s pamphlet and his discoveries. How reliable is his account? The foremost student of Seyxas is the Hispanist and book historian Clayton McCarl, who in a series of studies (McCarl Reference McCarl2014, Reference McCarl2018, Reference McCarl2020) has been drawn to the conclusion that Seyxas’ publications offer a mixture of fact and fiction. In Reference Seyxas y Lovera and McCarl2011 Dr McCarl published an edition of a work by Seyxas that had survived in manuscript in the collection of the Hispanic Society of America in New York City. The book, completed in 1693, was entitled Piratas y contrabandistas de ambas Indias, y estado presente de ellas [‘Pirates and smugglers of the East and West Indies, and the present state of those places]. In a subsequent article (McCarl, Reference McCarl2014), Dr McCarl analysed the bibliography that Seyxas had assembled, and discovered that more than 30 of the books listed by Seyxas seem not to have survived to the present day, nor are there any contemporary allusions to them. It is the case that in the Early Modern period a small number of books seem not to be extant today. The best known example is Shakespeare’s Love’s Labour’s Won, which was advertised in a quarto edition in 1603. But such losses are unusual, and the loss of 30 books on a single topic is hard to credit. The fact that many of the authors of books described by Seyxas have not been traced, and that the voyages and incidents that they describe are not otherwise attested, deepens the suspicion that Seyxas was a fabricator, and one with a modus operandi: he describes a voyage and describes a fictitious publication as his source.

In this context, Seyxas’s account of La Roche becomes suspect. The pamphlet that La Roche is said to have published in London does not survive, nor does La Roche’s diary, which is mentioned in Seyxas’s Theatro naval (Seyxas, Reference Seyxas y Lovera1688). In another article, McCarl notes that some details about La Roche are different in Seyxas’ two accounts. Whereas in the Descripcion geographica La Roche was an Anglo-French merchant living in London, in Piratas y contrabandistas he is described as an English merchant living in Cádiz. Text attached to a map prepared by Seyxas in 1703 offers more conflicting details: La Roche is said to be a pilot rather than a captain, and he is said to be French rather than English (McCarl, Reference McCarl2020).

Burney had mentioned two people other than La Roche for whom Seyxas was the sole authority. One was an Englishman called John Templemant, who is said to have sailed with Captain Narborough on his expedition to Patagonia, and to have published an account of the voyage in England in 1673. No such book seems to have existed, nor indeed did John Templemant. The other was Thomas Peche, who was said to have sailed into the Straits of Anian (the Bering Strait). Seyxas describes three meetings with Peche in Holland and explains that the account of Peche’s voyage was published in a pamphlet in 1679 in both French and English, in Holland, France and England. It is highly unlikely that a pamphlet printed in two languages and three countries cannot now be traced, nor indeed can its author.

It seems safe to conclude that the narratives of Francisco de Seyxas y Lovera consist in part of wholly invented characters, voyages and publications. La Roche’s pamphlet would seem to be one of Seyxas’s inventions, as indeed would La Roche himself. As La Roche seems never to have existed, he cannot be the person who discovered South Georgia. The argument about first discovery continues. The case for Gregorio Jerez as the first discoverer was revived at the time of the Falklands War by Rear-Admiral Laurio H. Destefani (Reference Destefani1982). Much archival research on the Jerez expedition has been conducted by Ricardo Capdevila (Reference Capdevila1988), who supplemented earlier work by Ernesto Fitte (Reference Fitte1962, Reference Fitte1968). The authors of a recent collection of essays discussing the peaceful recovery of the ‘Islas Malvinas, Georgias y Sándwich del Sur’ from a United Kingdom weakened by Brexit regard the case for Argentine sovereignty as overwhelming (Barbarán, Reference Barbarán2020). The debate will continue, but Antonio de la Roche should no longer be part of the argument.